Enhancing Crowd Image Analysis Through Facial Landmark Recognition
K. Soni Sharmila, R. Thriveni, T.Mukunda Priya · 2024
Face identification and recognition play a crucial role in various security applications, providing essential capabilities for our daily lives. The growing interest in computer vision stems from genuine concerns about public security in today's interconnected world. Facial features offer a straightforward means of distinguishing individual identities. Recent advancements in this field, fuelled by the availability of extensive datasets and advancements in deep learning techniques, have been noteworthy. Commonly, digital cameras are strategically positioned on rooftops of residences, buildings, and storefronts to monitor events and incidents on adjacent streets. Often, these cameras capture images featuring groups of people at a particular location. In the event of an unusual incident, security services must swiftly identify the responsible party. If these services possess images of potential suspects, a comparison and matching process can be initiated to identify the actual perpetrator. Typically, crowd analysis involves three key processing steps: preprocessing, object detection, and event/behaviour recognition. In this context, the goal is to recognize an individual from a crowd by utilizing a single image of the person to be identified. This process integrates face recognition and face detection modules, leveraging the advancements in these technologies to enhance security measures.